Typically, a solid state laser cavity contains a host material that is doped with a small amount of an activator ion. This ion can be pumped by a light source such as a flash lamp or more commonly, a diode laser of suitable frequency. The light from the pump is absorbed by the gain medium, i.e., the doped host, creating a population inversion that causes stimulated emission of coherent light. The output light can be in the form of continuous or pulsed emission.
While the gain medium can be the only crystal regime of a laser cavity, solid-state lasers often employ several single crystal regimes that serve a series of purposes including thermal management, mechanical strength, waveguiding, and the like. These regimes can occur in the form of a series of layers or films that have similar lattice structures and dimensions, but with slightly different chemical compositions where the different compositions reflect the different functions.
Thermal management is a significant matter particularly with regard to high-energy solid-state lasers. For instance, thermal lensing and surface distortion become a significant matter for higher power applications during which residual heat buildup can be localized in the active lasing cavity in a non-uniform fashion. This leads to irregular thermal expansion of the lattice and reduces the quality of the beam. Such localized heating also changes the refractive index of the material leading to gradient index (GRIN) behavior, which leads to beam defocusing. In addition, heating from the pump can induce cracking or other damage to the host crystal. This is a particularly sensitive issue because the damage is often induced at small defects at the crystal surface or the near subsurface that are especially sensitive to thermal damage. Another problem is distortion or cracking of any thin coating or film, such as Reflective or Anti Reflective coating applied to the surface of the crystal due to thermal deformation. Applied coatings are typically very thin and especially sensitive to surface damage. Many of these issues have been summarized in the technical literature (for example Armstrong et al. Optics Comm. 2000, 175, 201; McDonald et al. Optics Comm. 2000, 178, 383.), as well as the patent literature (see, e.g., U.S. Pat. No. 6,845,111 to Sumida, et al., U.S. Pat. No. 6,944,196 to Wittrock, and U.S. Pat. No. 5,761,233 to Bruesselbach, et al.).
One method for solving such thermal problems is to provide a moderately thick undoped host lattice regime attached directly to the doped lasing crystal. For example undoped regions of the host crystal have been placed at either end of a bar or rod laser to serve as endcaps. In this configuration the pump light will penetrate the undoped region of the crystal for a certain distance before it begins to get absorbed by the activator ions. The relatively thick endcap can dissipate heat buildup and minimize any distortion at the crystal face.
In thin disk lasers it is often desirable to have a thin layer region doped with activator ions on a thicker undoped region that serves as a substrate. The activator region is a thin disk so as to maximize surface contact with a heat sink. The undoped part of the crystal acts as a substrate to add mechanical strength and act as an interface with a heat sink or cooling zone. It may also be desirable to add an additional undoped layer of crystal to the opposite side of the cooling layer to minimize thermal damage to the reflective coatings. Such devices have been described in e.g., U.S. Pat. No. 6,347,109 to Beach, et al. and U.S. Pat. No. 6,834,070 to Zapata.
Lasing crystals have also been utilized in the form of planar or embedded waveguides. Waveguides require two zones with similar lattices whereby an internal portion, or core, contains an index of refraction that is larger than that of an outer portion, or cladding. Thus, total internal reflectance can be achieved and essentially all light that enters will exit at another end. This condition is most commonly exploited in fiber optics but can also work for other forms of matter including single crystals. It is particularly useful in that the waveguide can be thin enough to control the number of modes and the overall beam quality with little or no loss of power. In the case of crystalline planar waveguides, a thin layer of material (5-200 micron) doped with a small concentration of laser active ions can be adjacent to a thicker layer of undoped substrate (typically 50-500 micron). A further layer of undoped substrate can then be applied upon the doped layer to create a thin layer sandwiched between the two layers of thicker undoped material. If conditions are selected to ensure that that index of refraction of the thin activator layer is larger than the undoped regions, total internal reflection can be achieved and a waveguide created. Optimally it is possible to control the thickness of the activator layer so as to control the number and quality of the modes emitted.
Thermal control, strength, thin film designs and waveguiding are only a few examples of beneficial use of multiple crystal regimes in solid-state lasers. Whatever the purpose, the interface between adjacent regimes is of importance in forming such devices as it can have a large effect on overall beam quality. Moreover, if the resultant output beam is to be frequency manipulated through a non-linear process (for example second harmonic generation or optical parametric oscillation), it is important to have interfaces between adjacent crystal regimes with controlled lattice orientations to control polarization interaction with the pump light.
There are two general techniques presently in use to form heterogenous crystal devices. One method is direct bonding of the two different premade materials. Use of glues, fluxes or other bonding materials has been examined but is usually unacceptable due to degradation of the optical beam quality. Other direct bonding methods include pressure bonding, electrical potential fusion and other techniques, but these are often expensive, unreliable or otherwise not practical for scalable production of layers between 50-1000 microns thick. Such methods have been described in, e.g., U.S. Pat. Nos. 5,441,803, 5,563,899, 5,846,638, 6,025,060 and U.S. Patent Application Publication No. 2009/0041067.
A second method has been the growth of composite layers directly on a suitable substrate. Typically this has been accomplished through epitaxial growth in which one material acts as a substrate and a second material is deposited on the surface in a stepwise controlled manner. The grown layer adopts the general structural characteristics of the substrate (such as same lattice type and similar dimensions). Generally this process requires that the two materials have a similar structure type and reasonable crystal lattice match. In the case of solid-state laser devices, the use of gas phase epitaxial methods (molecular beam epitaxy, physical vapor deposition, MOCVD etc.) has not been suitable as gas phase methods are too slow to form the desired layer thickness.
Liquid phase epitaxy (LPE) as described by B. Ferrand, et al. (see, e.g., Opt. Mater. 11 (1999) 101-114; U.S. Pat. No. 6,973,115; EP Patent No. EP-A-0 653-82) has also been used. LPE employs high temperature fluxes to dissolve the substrate material and deposit the appropriate layers on the substrate seed via supersaturation. It typically employs molten salts that are usually mixtures of lead oxide and boron oxide or other metal oxides that melt between 1200° C. and 1600° C. and impart modest solubility to the desired layer material. Unfortunately, the LPE method often utilizes highly toxic lead-based solvents and requires very high temperature processing, leading to increased environmental danger and costs. Additionally, the formed boules must be treated to spin away flux and cleaned with nitric acid to remove any residual flux. Furthermore, the high temperature solvents often contaminate the resultant product with the flux and/or impurities in the flux.
Hydrothermal techniques, in which a temperature differential is developed to create a supersaturated solution leading to crystal growth on a seed, have been utilized for bulk single crystal growth (see, e.g., R. A. Laudise, J. W. Nielson, Solid State Phys. 12 (1961) 149-222), but are not known for use in forming heterogeneous materials. For example electronic grade quartz is grown commercially by the hydrothermal method. Other crystals, such as potassium titanyl phosphate (KTP) are grown by both flux and hydrothermal methods, and it is widely acknowledged by those familiar with the art that the hydrothermally grown products are of generally superior quality.
What are needed in the art are methods for forming solid state laser devices incorporating multiple crystal regimes that are more economical than previous methods. For example, a low temperature, facile process that can provide a monolithic heterogenous crystal for use in a laser cavity with no beam degradation problems due to interface between adjacent materials would be of great benefit.